Bipolar disorder is a common (1-2% of population), debilitating and potentially life-threatening psychiatric disorder that is characterized by recurrent, cyclic episodes of depression and (hypo)mania. Evidence from prior studies strongly implicates a genetic basis for the etiology of bipolar disorder. However, identification of the genetic factors that determine susceptibility has been elusive. Three recent genome-wide association (GWA) studies of bipolar disorder have produced few robust associations. These findings suggest that the genetic complexity of the disorder may require significantly larger sample sizes and a more comprehensive genetic platform, to detect the large number of associations with modest effects that may underlie the genetic basis of this disorder. This project is a collaboration of investigators at the Kaiser Permanente Division of Research and the University of California, San Francisco Institute for Human Genetics. The overall goal of the proposed study is to discover and characterize common genetic variants that may be associated with the risk of bipolar disorder, by conducting a GWA study of an ethnically diverse sample of 6,000 cases of bipolar disorder and 6,000 controls, ascertained among the members of the Kaiser Permanente Medical Care Plan, Northern California Region (KPNC). Cases and controls will be identified from longitudinal electronic medical records (EMR). To increase phenotypic homogeneity, the sampling frame will be limited to individuals with bipolar I disorder with multiple treatment episodes. Controls will be frequency-matched to cases on gender, current age, self- identified race-ethnicity, zip code, and length of membership in KPNC. Collection of saliva samples for obtaining DNA will be supported by the Kaiser Permanente Research Program on Genes, Environment and Health. Genetic analyses will be done at the UCSF Institute for Human Genetics and Genomics Core Facility. The specific aims for this study are to perform a GWA study of bipolar I disorder in the cases and controls described above using the Affymetrix 6.0 chip platform. We will analyze approximately nine hundred thousand single nucleotide polymorphisms (SNPs) and a similar number of copy number probes to assess potential case-control differences in genotype and haplotype frequencies, as well as copy number variation. Each ethnic subsample will be examined for population stratification, and consistency of results across ethnic groups will be evaluated. Formal significance will be determined by taking into account the large number of variants tested. Related phenotypes, such as the presence of psychotic symptoms or age at onset, will also be evaluated for potential associations with SNPs from the genome-wide scan. Collaborative studies with similar GWA study data will be undertaken to validate significant findings from our own and others' studies. Bipolar disorder is a relatively common psychiatric disorder that often involves impairment of occupational and social functioning and an increased risk of suicide. The causes of bipolar disorder are virtually unknown; however, a genetic basis for the disorder has been strongly implicated in previous studies. Genome-wide association studies in large, well-characterized samples such as the one proposed, can provide information about specific genetic factors that increase the risk of bipolar disorder, potentially leading to increased understanding of the underlying causes of the disorder, and to more reliable diagnoses and new treatments as well.